Skip to main content

Full text of "Decorative Symbolism of the Arapaho"

See other formats


This is a digital copy of a book that was preserved for generations on Hbrary shelves before it was carefully scanned by Google as part of a project 

to make the world's books discoverable online. 

It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject 

to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books 

are our gateways to the past, representing a wealth of history, culture and knowledge that's often difficult to discover. 

Marks, notations and other maiginalia present in the original volume will appear in this file - a reminder of this book's long journey from the 

publisher to a library and finally to you. 

Usage guidelines 

Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the 
public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing this resource, we liave taken steps to 
prevent abuse by commercial parties, including placing technical restrictions on automated querying. 
We also ask that you: 

+ Make non-commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for 
personal, non-commercial purposes. 

+ Refrain fivm automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine 
translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the 
use of public domain materials for these purposes and may be able to help. 

+ Maintain attributionTht GoogXt "watermark" you see on each file is essential for informing people about this project and helping them find 
additional materials through Google Book Search. Please do not remove it. 

+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just 
because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other 
countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of 
any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means it can be used in any manner 
anywhere in the world. Copyright infringement liabili^ can be quite severe. 

About Google Book Search 

Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers 
discover the world's books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web 

at |http : //books . google . com/| 


as \o303-/-yo 

$acbact) College 2.iiirars 












Submitted in Partial Fulfillment of the Requirements lor the 

Degree of Doctor of Philosophy, in the Faculty 

of Philosophy, Columbia University 

(Prom the American Anthropologist (N. s.), Vol. 3, April-June, 1901) 



190 1 


t -' - », . 

i J 




— «u^^^^^« 

■^.-K •■ 








,</ i f •' ^-^ 



The Arapaho, a tribe of Plains Indians belonging to the Al- 
gonquian stock, practise a form of art very similar in material, 
technique, and appearance to that of the other Plains tribes, of 
whom the Sioux are the best known. This art is in appearance 
almost altogether unrealistic, unpictorial, purely decorative. For 
the greater part it consists now of beadwork, which has nearly sup- 
planted the older style of embroidery in porcupine quills, plant 
fibers, and perhaps beads of aboriginal manufacture. The other 
products of this art are objects of skin or hide which are painted 
with geometrical designs. On the whole the decorative, geomet- 
ric character of Arapaho art is very marked. Almost all the 
lines are straight. The figures in embroidery are lines, bands, 
rectangles, rhombi, isosceles and rectangular triangles, figures 
composed of combinations of these, and circles. The designs 
painted on hide are composed of triangles and rectangles in differ- 
ent forms and combinations. 

On questioning the Indians it is found that many of these de- 
corative figures have a meaning. An equilateral triangle with the 
point downward may represent a heart ; with its point upward, a 
mountain. A figure consisting of five squares or rectangles in 
quincunx, the four outer ones touching the central one at the 
corners, is a representation of a turtle. A long stripe crossed by 
two short ones is a dragon-fly. A row of small squares at inter- 
vals represents tracks. Crosses and diamonds often signify stars. 
All this is in beadwork. In painted designs a flat isosceles triangle 

* Published by permission of the Trustees of the American Museum of Natural 


Mini I . , ^^B— BBSggg^llMI|L_|.MillB 

f • * --Hji.- - 

".<*^_V^* ;j.*r •'«t*_n**_»j»-?«~_«.*«»*«Bn 



'II . Ill 



It, JT^ 




kroeber] decorative SYMBOLISM OF THE ARAPAHO 309 

often represents a hill ; an acute isosceles triangle, a tent. Many 
other objects are similarly represented. 

An ornamental feature is the symmetrical duplication of most 
designs. Bags, pouches, skins, moccasins, cases, and other objects 
are ornamented by being treated as a decorative field within 
which the designs are symmetrically doubled, or even more nu- 
merously repeated. Thus a moccasin, if decorated with the symbol 
of a mountain on the outer side of the heel, has the same symbol 
also on the opposite inner side of the heel. Another purely 
ornamental feature of this art is repetition of a single figure to 
form a pattern. A stripe is often the representation of a path. 
This symbol is sometimes used singly, standing alone ; sometimes 
it occurs double, owing to the tendency just mentioned, toward 
symmetry ; and sometimes it is found in a pattern that may be 
described as a many-colored, drawn-out (i.e., rectangular, not 
square) checker-board, in which each rectangle or short stripe, 
whatever its color, still represents a path. 

This strongly-marked decorative character of Arapaho art, 
however, is accompanied by a realistic tendency of such develop- 
ment as at first acquaintance would not be suspected by a civilized 
person. Several figures connected in meaning may be put upon 
one object and thus produce something approximating a picture 
containing composition. When as many as ten or a dozen sym- 
bols having reference to each other are combined, a story can al- 
most be told by them. In this way the stiff embroideries on a 
moccasin or the geometric paintings on a bag may represent the 
hunting of buffalo, the acquisition of supernatural power by a 
shaman, a landscape or map, a dream, personal experiences, or a 
# Arapaho art thus is at the same time imitative or significant, 

and decorative. Can the origin of this art be <i^etermined ? 

Since Arapaho art consists of the intimate fusion of symbolism 
and decoration, two theories as to its origin are possible. Either 
of its two elements may be the original. The Indians may have 

310 AMERICAN ANTHROPOLOGIST [n. s., 3. 1901 

begun with realism, drawing or working lifelike forms in their 
art ; then, however, the obstacles inherent in the material as- 
serted themselves, or the well-established tendency toward sym- 
metry and repetition into a pattern came out, or perhaps other 
causes were influential, until the early imitative representations 
became abbreviated into the conventional decorations that have 
been described. Or it is possible that the Indians began with 
mere ornaments. Perhaps even these were not originally orna- 
ments but peculiarities of construction of purely useful articles, 
which technical peculiarities were later considered beautifying and 
developed into pure ornaments. At any rate, whatever their own 
origin, decorations may in the past have existed/^r se ; later, some 
conventional ornament may have accidentally suggested a natural 
object, whereupon it was modified to resemble this object more 
closely ; the same process occurred with other ornaments ; until 
finally a whole system of symbolism was added to the older 
system of decoration. The first of these theories is that original 
pictures were conventionalized into decorative symbolism ; the 
other theory is that original ornament was expanded into symbolic 
decoration. These are the logically possible explanations of the 
origin of Arapaho art because we recognize in it two factors, the 
realistic-symbolic and the decorative-technical. 

Let us see if either of these theories can be rendered through 
the evidence of fact actually certain or at least probable. 

One of the most frequent embroidered designs on Arapaho 
moccasins consists, in its simplest form, of a stripe or band which 
runs from the instep to the toe. This decorative motive takes 
varied forms, of more or less elaborateness. The following are 
a number of moccasins with this type of ornament. 

One moccasin* (PL V, fig. i, catalogue number g \ q ) is em- 
broidered merely with a stripe from the instep to the toe. This 
stripe of beadwork consists of a number of bars or lengthened 

^ The Arapaho objects described in the course of this article are in the American 
Museum of Natural History. Their catalogue numbers are given in parentheses. 


kroeber] decora TIVE SYMBOLISM OF THE ARAPAHO 3 1 1 

rectangles of different colors. No information was obtained as to 
the meaning of the design on this specimen. 

Another specimen (PI. v, fig. 2, cat. no. -^y^) has a similar 
stripe, about an inch in width, running from instep to toe, and 
composed of bars or small stripes of six different colors. The dis-^ 
tribution of these colors is not like that in the last described speci- 
men, but the pattern and the idea of color arrangement are iden- 
tical. On this moccasin, however, there is one additional piece of 
embroidery, a narrow stripe across the instep, that is, transverse to 
the main stripe and touching it at its upper end. The large stripe 
as a whole, the smaller bars separately, and the transverse stripe 
all represent buffalo paths. 

A third specimen (PI. V, fig. 3, cat. no. -^^ also has a 
stripe from instep to toe. This is white, except for a rectangular 
green portion in the middle. At the two ends of this green part 
of the stripe are two dark-blue (=black) marks, which are approxi- 
mately triangular. Across the instep we again find a narrow 
transverse stripe. This represents a bow. The main longitudi- 
nal stripe represents a buffalo path. Its green rectangular portion 
is a buffalo. The black marks are arrowpoints shot into it. Small 
projections on these marks, which render them not really quite 
triangular, represent the barbs of the arrowheads. 

Another moccasin (PI. V, fig. 4, cat. no. y^Jx) again has the 
longitudinal stripe. This represents a path, probably with impli- 
cation of the path traveled by the wearer of the moccasin. The 
major part of this stripe is white, but portions are beaded in dark- 
tlue (=black), red, and grayish-blue. These colors denote respec- 
tively night, day, and hazy atmosphere. On the white stripe are 
also two curious symbols, which are said to signify sunrise or going 
over a mountain. A narrow transverse stripe is found in this 
specimen also ; but instead of being contiguous to the end of the 
main stripe, as on the last two moccasins, it is cut in two by it, so 
that it exists only in two fragments, one on each side of the large 
stripe. These two small bars represent insects that are desired to 

312 AMERICAN ANTHROPOLOGIST [n. s., 3, 1901 

be out of the path, beside it, instead of being where the moccasin 
will travel in the path. 

Another specimen (PL V, fig. 5, cat. no. -^^-^ has the main 
stripe down the foot slightly modified in that it tapers a little 
toward the toe. In arrangement of colors, this moccasin resem- 
bles closely the second one described. In all the specimens just 
discussed, except the last, the bars of which the main stripe con- 
sists are arranged in three groups. In this moccasin this triple 
division of the stripe also exists. Moreover, in the middle section 
of this stripe there is a green rectangle, and in contact with this a 
small dark-blue mark approximately triangular in shape. These 
two symbols are very like the representations of the buffalo and 
arrowpoints on the moccasin above described as symbolic of the 
buffalo hunt. Unfortunately it is not known whether the design 
on the present specimen had any meaning. So far, accordingly, 
this moccasin agrees closely with those previously examined. It 
is further like them in possessing a narrow, transverse stripe of 
beadwork at the instep. But a totally new feature is found in two 
small bars that start from the ends of the transverse stripe. They 
are parallel to the main central longitudinal stripe, but very much 

In all the preceding specimens but one (fig. 4), the large stripe 
consists of three sections. In the exceptional specimen the upper 
third or fourth of the stripe is of one ground color, the remainder 
all of another ground color. Such an arrangement is also found 
in another specimen (PI. V, fig. 6, cat. no. -jpjVr)' ^^^ smaller 
portion of the stripe is white, the longer part is blue with a pattern 
imposed upon it. Nothing is known of the significance of any 
part of this design. The two small bars are present, as in the last 
specimen, and repeat the markings of the large stripe in simplified 
form. But the transverse stripe at the instep is missing. 

Still another moccasin has its stripe divided into a short upper 
and a long lower portion of different colors (PI. V, fig. 7, cat. no. 
■j'y'j). As in the last specimen, there are two small bars 

kroeber] decora TI VE S YMBOLISM OF THE ARAPAHO 3 1 3 

parallel to the central stripe and repeating its design, and the 
transverse stripe is again absent. The stripes and bars all repre- 
sent buffalo paths. In certain parts of the stripes are small squares 
colored light blue ; these represent buffalo tracks. 

The last specimen of this series (PI. V, fig. 8, cat. no. -^-^ 
has the main central stripe, the transverse stripe at the instep, 
and the small bars repeating the markings of the large stripe. In 
addition to these three decorative devices that are found in pre- 
vious specimens, it possesses a fourth one that is new. The cen- 
tral longitudinal stripe (slightly constricted toward its middle) is 
bisected by a duplicate of itself running transversely. These two 
stripes thus form a cross. This cross represents the morning star, 
the variety of colors upon it denoting the variety of colors the 
star appears to assume. The transverse stripe at the instep rep- 
resents the sky or horizon. The two small bars are said to be the 
twinkling of the star as it rises, in other words its rays.* 

The symbolism of some of these designs is elaborate. The 
representation of the buffalo in his path shot by arrows from the 
hunter's bow is coherent and neatly compact. We do not know 
whether it is a commemoration of a particular event or the ex- 
pression of a wish for plenty of food, but in either case it has 
pictographic function. In fact, it is a pictograph, except for the 
fact that its geometric form renders it illegible for any one but its 
writer. The star-moccasin is also a pictograph in an ornamental 

The conventionality of the decoration seems to have reached 
an equally strong development. It is apparent that the large 
stripe from instep to toe is the fundamental motive of this style 

^ Some of these moccasins, it will have been noticed, are without known symbolism. 
This is due merely to their having been collected without inquiry being made as to 
the significance of their designs. Consequently, to judge from analogy, it is more 
probable that they do have meaning than that they really lack it. 

^ Even in true pictographs free from decorative limitations and therefore drawn 
with the greatest realistic fidelity of which the Indian is capable, the symbols for the 
morning-star, the horizon, and rays of light are the same as those on this moccasin — a 
cross, a horizontal line, and vertical or sloping lines. 

314 AMERICAN ANTHROPOLOGIST [n. s., 3, 1901 

of ornamentation. All the other motives are also stripes, and 
even of these there are only two (the transverse stripe and the 
two short bars), except in the one .morning-star moccasin where 
the basal element is introduced in a new position as a fourth 
decorative motive. 

In short, in these moccasins the tendency to realistic symbol- 
ism and the tendency to decorative conventionalism are clearly 
about in equilibrium. Hence we cannot fairly say that either of 
these tendencies is the older and original. If one concentrates 
his attention on the symbolism, or happens to be temperamentally 
more interested in it, he is very likely to see it more abundantly 
than the decoration, to be more impressed by it, to consider the 
entire present art as merely corrupted or abbreviated symbolism, 
and to advance as an explanation of the origin and development 
of these designs the theory of conventionalized realism. But if 
one thinks more of the decoration as such, or if one's mind runs 
naturally toward the ornamental and technical, he will probably 
notice mostly this side, regard the significations of markings as 
trivial and irrelevant additions that may be ignored, and finally 
champion the theory of expanded decoration. With the one bias 
we are so overwhelmingly aware of the almost pictographic coher- 
ence in the buffalo-hunt moccasin, that we believe that pictures 
of such topics must have given rise to the present form. With 
the other bias the conventionality of the pattern that possesses 
this buffalo-hunt significance is so impressive that we come to 
think that decorative motives x)f just such persistence as this must 
have been the origin of the present form. A first investigator is 
so struck with the enormous difference of meaning between the 
ordinary path-stripe moccasins and the morning-star-cross mocca- 
sin that he cannot believe they had a common source ; each must 
have sprung from a picture, which was as different from the other 
as the objects represented are different. A second observer is so 
impressed by the fact that the morning-star moccasin with four 
decorative elements differs less from some of the buffalo-path 

kroeber] decora TI VE S YMBOLISM OF THE A RAP A HO 3 1 5 

moccasins than many of these with from one to three decorative 
elements differ from each other, that he thinks that all these de- 
signs, however variable their superficial meanings, must have 
originated in one typical ornamental form. 

Both these explanations are thus, in the case of these moccasin- 
designs, not only logically possible, but they are very naturally 
believed and advanced as the result of certain mental predis- 
positions. But if we try to remain free from any such inclinations 
of mind, and if we remember how strongly developed and inti- 
mately fused are both the tendencies, we must come to the con- 
clusion that, because symbolism and decoration balance each 
other, the two theories of conventionalized realism and expanded 
ornament, though logically admissible, are actually untenable. 
Rather it seems likely, since the two tendencies are vigorous, and 
combined, that they are both well established, old, and long in 
close union ; so that formerly designs on Arapaho moccasins, 
though perhaps ruder than now, were of the same general 
character, both symbolically and decoratively, as those we know. 

Let us consider a second style of moccasin. Whereas in those 
just discussed the fundamental element of the embroidery was 
the longitudinal stripe, it now is a border running all around the 
foot just above the sole. In one particular specimen illustrated 
(PI. VI, fig. I, cat. no. ^ttV'S") there is besides this border of bead- 
work a series of lines of quillwork filling the large space on the 
front of the moccasin, but as this is embroidery of a different ma- 
terial and appearance, we can disregard it in the present considera- 
tion and confine our attention to the ornamentation consisting of 
the border. It should be added that in addition to the border 
there is the narrow stripe across the instep. 

In a second specimen (PI. VI, fig. 2, cat. no. -^) there is 
besides the border and the transverse stripe, the large longitudinal 
stripe with which we have become familiar. As previously, this 
signifies paths. 

A third specimen (PI. VI, fig. 3, cat. no. ^^-^^ has the border. 

3^6 AMERICAN ANTHROPOLOGIST [n. s., 3, 1901 

the large longitudinal stripe, and the two small bars at its upper 
end, but lacks the transverse instep-stripe. On the central stripe 
are two representations of birds, but there is no information as to 
the meaning of the design. 

It is evident that in these last two moccasins there is a com- 
bination of the stripe motive with the border motive. 

In another specimen (PI. vi, fig. 4, cat. no. Vi?V)> ^^ whose 
symbolism we are ignorant, the longitudinal stripe is continued 
farther than previously, so that it meets the border. The stripe 
is not solidly embroidered : its edges are beadwork, but its interior 
is left open and merely painted red. 

In any moccasin of this design there is left a blank space on 
each side of the foot. This is the area enclosed by the stripe, the 
border, and the transverse instep-stripe. It has the shape of a 
pointed right-angle triangle whose hypothenuse instead of being 
straight is convex. These two triangular or horn-shaped areas 
occur in another moccasin (PL vi, fig. 5, cat. no. ^]^y). The 
border, stripe, and transverse stripe are all white. The two 
enclosed areas are half covered with a checker-board design in 
several colors, which is said to represent buffalo-gut. This check- 
er-board embroidery also extends around the heel. 

If, now, this half-open checker-board work were replaced by 
solid beading, we should have a moccasin completely covered 
with beadwork. Such specimens occur in abundance. In one 
(PI. VI, fig. 6, cat. no. ylS^-g-), whose groundwork is white, the two 
triangular areas taken together represent buffalo horns. The buf- 
falo trample the ground ; this is represented by the coloring of the 
two areas. One is red, which denotes the soil, or bare earth ; the 
other is green, which denotes vegetation or grass-covered earth. 

A child's moccasin, also solidly beaded (cat. no. tIjt)* has 
as usual a groundwork of white. The two triangular areas are 
green, and represent horse ears — a symbol of good fortune and 
future wealth. Between them, the central stripe, slightly modified, 
represents a lizard. 

kroeber] decora TIVE S YMBOLISM OF THE ARA PA HO 3 1 / 

A last moccasin (cat. no. \\\i ^ is solidly beaded in white. 
The two triangular marks are banded dark-blue and white, and 
represent fish. 

In these last cases, in fact in most fully beaded moccasins, the 
decorative elements of border, stripe, transverse stripe, and 
triangular area are still visible in the embroidery ; even though 
they often become identical in color and are not distinguished in 
the design, they are used technically. 

If we follow the transition from the merely bordered moccasin 
to the solidly beaded one, and see the same technical or decora- 
tive features persisting in all parts of the series from the simplest 
to the most highly developed form, the ornamental nature of 
these productions is striking and their decorative origin seems 
probable. If we consider the realistic representation of, for in- 
stance, the buffalo horn, and the pretty symbolism of its coloring, 
the realistic origin of these decorations seems very hard to dis- 
believe. Of course there is no reason for leaping at either of these 
conclusions. Neither phase of this art must be ignored, but both 


recognized. It is necessary to be aware both of the strong orna- 
mental tendency influenced by symbolism and the symbolic 
tendency modified by ornamental system. 

So far as these moccasins are concerned, it accordingly seems 
impossible to determine with certainty how the symbolic decora- 
tion originated. 

Parfleches and bags of rawhide made by the Arapaho are 
painted on the front with designs that cover most of the surface. 
The back or bottom is sometimes left blank, or may have from 
six to ten straight lines (or narrow stripes) painted transversely 
across (fig. 49). These lines on the bottom usually represent 
roads or rivers. All parfleches are perforated in front to allow of 
being fastened with thongs. Occasionally, however, a cautious 
person winds a rope a number of times around his bag, in order to 
tie it up more securely. On one parfleche seen by the writer such 
transverse lines were painted across the bottom. The owner and 



[n. s., 3, 1901 

maker declared that they represented a rope passing over the sur- 
face of the back several times for fastening the bag. She showed 
another parileche in her possession which was actually thus tied. 
In this case the markings may appear to be an instance of the 
survival, as a decoration, of an atrophied useful feature: first 
ropes were regularly wound around the parfleche to fasten it, 
then these were left off but were represented by painting. This 
technical-ornamental theory seems at first glance to offer the true 

explanation o f 
the origin of 
these lines on the 
back of all raw- 
hide bags. But 
a moment's con- 
sideration shows 
that it is also 
within reason to 
believe the oppo- 
site: we can de- 
clare that the lines originated from attempts at representing 
rivers or roads, but that in this case the maker of the bag was 
struck by the resemblance of the lines to a rope as it was oc- 
casionally used, and then gave the new signification of rope to 
what really were conventionalized representations of rivers or 

So here again we have two explanations (there may be still 
more) that are plausible, while neither can be proved conclusively. 
As soon as we go beyond the description of existing circumstances 
into the inquiry of origin, we enter the realm of uncertainty, of 
irrefutable doubts. 

A peculiar Arapaho medicine-case shows unusual symbolism. 
The design painted on this is shown, spread out flat, in fig. 50 
(cat. no. VVV)* ^^^ ornamentation, which is less geometric than in 
most specimens of painting, represents the acquisition of super- 

FiG. 49 — Marking on Arapaho bag. 


kroeber] decora TIVE SYMBOLISM OF THE ARAPAHO 3 19 

natural power. Below, on the right side, is the sweat-house into 
which the owner and maker of the case went before beginning his 
fast to acquire supernatural power. This ornament also represents 
a small mound in front of 
the sweat-house, on which a 
buffalo skull is lying. The 
fish-tail ornament just above 
this is the mountain on which 
the man fasted, and hence 
also represents himself. To 
the right of this, the crescent- 
topped design is "the over- 
seer*' (the sun), also called 
" the one that lights.'' The 

pedestal or stalk of this figure Fig. 50— Design on Arapaho medicine-case. 

represents "information " 

(supernatural power) flowing down from this being to the earth 
(the horizontal line). At the extreme left, the same design is a 
representation of himself after he had acquired information and 
power ; and to the right of this, the fish-tail ornament now repre- 
sents this very medicine-case. But the case is made of buffalo- 
hide, and his supernatural power consisted largely in control of 
the buffalo; therefore this same symbol also denotes buffalo. 
Below, on the left, is the sweat-house into which he went after 
his fast. 

We have here an example of highly-developed symbolism. It 
might seem that when so long a story is told and so much abstract 
information is conveyed, the ideographic design must have arisen 
directly from the attempt of the artist to express his meaning, i. 
e., that the design is quasi-realistic in origin. But there is another 
medicine-case (fig. 51^, cat. no. -^-^ with similar ornamentation 
(about whose signification we unfortunately have no information). 
The resemblance of the two designs is great. One consists of an 
alternating arrangement of two symbols, both forked, the other of 




■ 3, 1901 

an alternating arrangement of these two symbols with a tliird, the 
semicircle, added. Some Arapaho say that this style of case was 
used by a powerful medicine-man and his followers or scholars ; but 

it is uncertain 
whether this man 
invented the de- 
sign or used an 
already existing 
one. It is doubt- 
ful whether even 
the symbolism 
was originated 
by this man or 
was similar to an 
earlier current of 
symbolism. The 
most usual orna- 
m en tat ion on 
Arapaho medi- 
cine-cases i s a 
pattern of tents 
(fig. s lb, cat. 
no. VW) o"" a 
combination o f 
triangles and 
diamonds similar to that painted on parfleches. 

So here again there is pictographic symbolism fused with a 
more or less conventional decoration, and it is impossible to say 
whether the symbolism or the decoration is the older and original. 
Small paint-bags — buckskin pouches to hold body-paint — are 
in general use among the Arapaho. Some of these represent half 
of a double-ended fringed saddle-bag. The rest all represent 
small animals, such as the beaver, lizard, rat, fish, mussel, horned 
toad, and frog. The opening represents the animal's mouth, two 



Strings that serve to tie up the opening are its forelegs, a loose 
flap at the end may be the tail, the pouch itself is the body, and 
other parts are indicated, as there is need, by beadwork, strings, 
or attached ornaments. The resemblance to the animal repre- 
sented is often detailed, but never accurate, being ideographic 
rather than visual, in keeping with all the symbolism of this art. 
It is generally impossible to recognize what species of animal is 
meant, and only the maker knows this. 

One pouch represents both a beaver and a fish (fig. S2«, cat. 
no. -^^), according to information given by its owner. When it 

is regarded as a beaver, both pairs of strings are legs, and the 
scallops or notches at the opening are the prominent teeth. ^ A 
design in beadwork on the pouch, which represents a stream with 
a dam and beaver-huts, also refers to this signification. When a 
fish is meant to be represented, the upper pair of strings are the 
barbels, the lower pair the pectoral fins. The fish-signification is 
strengthened by a rough line of beads at the edge of the pouch, 
which are interpreted as fish-scales. 

A very similar pouch represents a lizard (fig. 52^, cat. no. -^^. 
Mouth, body, legs, and tail are represented in the conventional 
manner by opening, pouch, strings, and attached flap. 

322 AMERICAN ANTHROPOLOGIST [n. s., 3, 1901 

A pouch that lacks the long flap represents a frog (fig. 52^, 
cat. no. ^jV). Two long strings indicate the frog's hind legs. A 
fringe at the bottom represents the grass in which it is sitting. 
A design in beadwork on this pouch denotes the shoulder and 
hip joints of the frog, and the food in its stomach. 

Another pouch (fig. 52^, cat. no. /Z^) differs in shape from 
this one only in lacking the two longer strings. It represents half 
a saddle-bag. 

The realistic tendency manifested in the animal symbolism of 
these pouches is undeniable. A conventional, formal, decorative 
tendency is evident in the close similarity between the frog-pouch 
and the saddle-bag pouch,, and between the beaver-pouch and the 
lizard-pouch. Both the tendencies come to light in the pouch 
with the curious double signification. 

Some of the Arapaho say that at the beginning of the world, 
when the first men, their ancestors, obtained paint, they had only 
the skins of small animals to use for paint-bags, and that this is the 
origin of the animal symbolism of the present-day paint-pouches. 

It is necessary not to be misled into a belief of this origin and 
development on the authority of the Indians. Their authority on 
such a point is absolutely valueless. They believe that the time 
when the first men obtained paint-bags was four hundred years 
ago, just after the formation of the world by a solitary mythic 
being floating on the water, and after a female whirlwind enlarged 
the minute earth by circling about it. Like all American savages 
they are almost completely without historical sense or knowledge. 
Occasionally a striking event may be remembered in a distorted 
form for a century or two, but on the whole, whatever of actual 
occurrence is retained in their tales is inextricably blended with 
mythic and supernatural elements. We have no right to reject 
the greatest part of their creation myth as so absurdly impossible 
that it would enter no one's mind to accept it as true, and at the 
same time to select here and there a point that is within the 
limits of possibility and proclaim it as historical and reliable. 


The mythic and historical elements in primitive legends are not 
simply mixed together so that they can be distinguished and 
separated, but they are both equally wonderful and equally true 
for the savage. No myth can be intrepreted into history by mere 
elimination of its supernatural portions : it must be rejected in 
toto. Even though it may be founded on a basis of actuality — 
and this must often be the case — it is altogether myth. In law, 
and exact science, and wherever evidence is judged, an account 
that is in great part manifestly absurd or palpably impossible is 
not accepted as true after the impossibilities have been subtracted,, 
but is disregarded as a whole. So, too, it is necessary to attach 
no importance to the statement of the Arapaho as to the origin 
of these paint-bags. 

We have considered several forms of Arapaho art — various ob- 
jects, various styles, and various materials and techniques. In all 
cases we have found a well-developed symbolism and a conven- 
tional decoration. The symbolism and the decoration exist not 
side by side but in each other. It has been easy to manufacture 
explanations of the origin of this art that are plausible theories. 
But as soon as we are open to recognize all possibilities, such 
theories are seen to arise from our opinions and methods of inter- 
pretation, and to be unsubstantiable by fact. Therefore we can 
describe Arapaho art, we can characterize it, and distinguish its 
various coexisting tendencies. We can even, to a certain extent, 
enter into the spirit of the people who practise it, and understand 
(1. e., feel) their mental workings. We cannot in fairness lay claim 
to knowing the cause or origin of this art, nor can we hope to as- 
certain its cause and origin by studying its products. 

In the art of other primitive races conditions very much re- 
semble those just discussed. Everywhere art is conventionalized, 
under the influence of a definite style. Practically everywhere 
also it is decorative. This is obviously true of such high arts as 
those of the Japanese and Chinese. It is true also of Greek 
sculpture and of Renaissance paintings : though in our modern 

324 AMERICAN ANTHROPOLOGIST [n. s., 3, 1901 

civilization we are in the habit of regarding the products of these 
arts detachedly, and enjoy them as if they were complete in them- 
selves, yet every one is aware that the intent to decorate always 
accompanied the conception and execution of the classic and 
Italian masterpieces. Even so strenuously realistic an art as 
modern impressionism is unable to free itself totally from the re- 
proach of being ornamental ; for whatever the purpose of the 
artist, the owner of such a picture has almost certainly secured it 
for the purpose, ostensible at least, of decorating a vacant wall. 
In primitive civilizations, the combination of the imitative and 
decorative tendencies is of course much greater. With very few 
exceptions, such as in some Eskimo tribes, the realistic, representa- 
tive impulse is thoroughly impressed and influenced by the highly 
conventional style ; and in all cases this conventional style is dec- 
orative. Correspondingly, most primitive decoration, no matter 
how geometric or simple, has significance and thus is, visually or 
ideographically, realistic. This is a fact that has not become known 
until recently, because until lately savages were rarely questioned 
thoro«ghly.* Accordingly the main characteristic of Arapaho 
art, its fusion (which is more truly an undifferentiation) of the 
realistic and decorative tendencies, is also the characteristic of all 
primitive art. 

In Brazil we know of tribes whose painted and incised de- 
signs, which are exceedingly simple and geometrical and usually 
in patterns, are all significant. Diamonds whose corners are 
slightly filled in are rhomboidally shaped fishes ; a pattern of flat 
isosceles triangles stood up on end is hanging bats, and so on. 
There are also other representations of the same animals that are 
slightly more realistic. The same tribes use pots of oval shape 
with half a dozen variously shaped projections at the rim. The 

* The scarcely suspected inherence of realistic significance in primitive ornament 
has been independently demonstrated from California, British Columbia, Central 
America, Brazil, Mississippi valley, Siberia, Indo China, Borneo, New Guinea, Aus- 
tralia, and Polynesia, in arts as diverse as pottery, weaving, carving, basketry, draw- 
ing, and painting. 

kroeber] decora TIVE SYMBOLISM OF THE ARAPAHO 325 

whole vessel represents an animal, the projections being roughly 
modeled into head, tail, and limbs. Birds, bats, mammals, rep- 
tiles, and invertebrates are indicated by very slight modifications. 
A civilized person unacquainted with the mode of sight and 
thought of the Brazilian aborigines might very readily mistake a 
bird-pot for a mammal-pot, and so on. 

In Central Australia bullroarers and other objects are deco- 
rated with incised lines. These consist of concentric circles, bands 
of parallel lines, concentric arcs or curves, and rows of dots or 
small marks. The ornamentation is not symmetrical, nor even 
regular ; it appears random and rude. Yet in general character 
these decorated bullroarers resemble each other closely. It has 
been found that the designs are all ideographic, though the total 
range of significance is apparently not very wide. Similar marks 
may on different objects mean things as different as trees, frogs, 
eggSj or intestines. It is interesting to note that while this art is 
remarkably crude and unformed both as regular ornamentation 
and as an attempt to represent objects accurately, it contains a 
system of realistic expression as well as a system of decoration, 
both of which are conventionalized — or rather, the union of which 
is a convention. 

The remarkable art of the North Pacific coast of America is 
certainly one of the most stylistic and conventionalized in the 
world, while its realistic character is sufficiently marked to give 
no one room to doubt its presence. Its decorative tendency is so 
strong that, in obedience to its demands, an animal that is being 
represented may be cut into parts which are then arranged as 
suits the requirements of the decoration and not as they are in 
nature. The chest of an animal may be put over its head, and 
the tail below ; two opposite sides of an animal, which are of 
course invisible at the same time, will be represented, in order to 
meet the strong demand for symmetry. The chief decorative 
motive of all this art is an oblong figure whose corners are 
rounded and whose sides are very slightly convex, the upper 

326 AMERICAN ANTHROPOLOGIST [n. s., 3, 1901 

long edge generally curving the most. Almost everything that 
is represented is brought into this shape or some modification 
of it. Heads, eyes, mouths, ears, joints, tails, fins, are usually 
of this shape ; the whole decorative field itself often is the 
same; and in such cases the remaining portions occupied by 
unenumerated parts, such as back, belly, and wings, are almost 
necessarily of the same shape. Eyes and faces appear every- 
where — on representations of joints, of the chest, of dorsal fins, 
of hands, in vacant spaces — and their shape is regularly the 
ornamental one described. Yet with this remarkably strong 
decorative tendency pervading and deeply influencing every repre- 
sentation, all examples of decorative art from this region are rec- 
ognizably realistic in intent and often in execution. There is no 
geometrical ornament that one might take to be meaningless. 
In short, on the North Pacific coast of America all decoration is 
realistic and all realism is decorative. 

It is of course impossible to prove by selected examples such as 
these that all primitive art consists of the combination of represen- 
tative realism and ornamental conventionalism. But that such is 
the fact, that this undifferentiation continues often into a higher 
civilization, must be obvious to any one familiar with primitive art. 
This fusion of two differing tendencies is not merely a frequent 
or widely distributed occurrence, as are a great many special 
ethnic phenomena, such as circumcision or doctoring by sucking 
or angularity of ornament, but this fusion is a rule practically 
without exceptions. It is universal because it is necessary. Both 
the representative tendency and the decorative tendency are 
deep rooted in the human mind, so that it must be virtually im- 
possible to suppress them for any length of time or among any 
considerable number of men. At times, indeed, as in European 
civilization, the two tendencies become more separated : our 'wall- 
papers are chiefly ornamental, our oil paintings chiefly realistic. 
But a glance at the past and present races of the world shows 
that this condition is exceptional, just as a civilization of the 


kroeber] decorative SYMBOLISM OF THE ARAPAHO 327 

extremity of ours is exceptional* The more primitive a people 
is, we may say, the more intimately fused in its art will these two 
tendencies be, though, as there is no absolute or fixable scale of 
primitiveness and civilization, this rule cannot be applied to 
special cases but merely tends to be true. Other tendencies also 
are still combined with these two in a sufficiently early and rude 
condition of society. The symbolism of the Arapaho is as ideo- 
graphic as it is realistic, and is as much a primitive method of 
writing as it is of artistic representation. The Australian bull- 
roarers referred to are, in addition to other things, very primitive 
maps or charts ; so that they are the products of diagrammatic, 
graphic, visually artistic, and decorative tendencies or activities 
still undifferentiated — all this in addition to their still more 
marked religious functions. Of course it is possible for a race to 
over-develop one of several related tendencies at the expense of 
others. To a certain degree this does happen in all races, and is 
what makes the difference between them. But every culture 
must contain among its motive forces more or less of every ten- 
dency, because the tendencies are in the human mind and hence 
ineradicable. These many tendencies are on the whole less dif- 
ferentiated in more primitive conditions of society. Hence all 
art, and especially primitive art, contains the combination at least 
of representative and decorative tendencies, perhaps of others.* 

^ The differentiation here and previously spoken of as accompanying or constituting 
evolution in civilization is at once too important and universal a matter to be proved 
here in a few incidental words, and too obvious to require it. A striking example of 
this differentiation is found in the mythology of our more primitive forefathers, 
in place of which, and more or less developed from which, we have products as 
different as romantic novels, fundamental scientific theories, and the doctrinary 
beliefs of our religions. There is no intention, however, of implying here by dif- 
ferentiation a continuing separation. Where in a savage tribe every man, though 
in somewhat varying degree, is hunter, warrior, participant in government, sha- 
man, artist, and myth-maker, a higher nation has its separate politicians, soldiers, 
food-producers, physicians, poets, and so on ; but though the tendencies have in 
this transition differentiated, and have far more than formerly become specialized 
in individuals, yet they exist only in the culture as a whole : in this, the only 
true unit, i. e., the only organic entity, they are all combined. For instance, our 

328 AMERICAN ANTHROPOLOGIST [n. s., 3, 1901 

The invariable method of explaining the origin of an art has 
been to select that one of its tendencies which was the most marked 
or appeared so to the investigator, to imagine the products of this 
tendency in its most isolated and pure form, and to pronounce 
these the original state of the art. An observer is struck by the 
fact that in a certain primitive art many ornamental features co- 
incide with technical ones that are present for practical reasons. 
He concludes that the technical-practical tendency which he has 
discovered among the decoration, is the original unmixed impulse 
that caused the art. Or he may become aware through inquiry 
or study of the fact that geometric ornament in an art has realis- 
tic significance. The realism impresses him ; true, it is now modi- 
fied and corrupt, but that only proves to him that originally it 
was pure. Ergo, this art began with representative pictures. 
Such has been the only method of explanation, however much 
the actual results in different cases differed. No other method of 
ascertaining or explaining the origin of a primitive art whose 
history we lack, is even possible. 

This method has the fundamental fault that it presupposes 
tendencies to have existed more unmixedly and separately at 
some former time than at present. In reality they must in all 
cases have been in the near past very much as now and in the 
very remote past more mixed or mutually undifferentiated. Thus 
we have seen that Arapaho art must some time ago have been 
very much as now. What it was still earlier we know even less 
definitely, but we cannot doubt that its spirit must have been 
similar. Different objects may then have been represented, other 

present-day science could not have arisen nor could it exist without modern industrialism, 
and this is equally dependent on science. Our literature is absolutely and intimately 
interwoven with our social conditions, not so much in that poets and novelists actually 
describe these, but in that the emotions and ideas which form the content of their writing 
are the typical emotions and ideas accompanying our social circumstances. In pro- 
portion with the differentiation of tendencies in evolution proceed their combination 
and recombination. Very analogously, a mammal is i2fc more highly differentiated than 
a jellyfish, but none the less are its various organs interdependent and itself a distinct 
organic unity. 

kroeber] decora TIVE SYMBOLISM OF THE ARAPAHO 329 

ornamental motives employed in other materials, but even then 
there certainly was the combination of ideographic symbolism with 
crude, heavy decoration. As we go farther backward in time, we 
can be sure that the details of the art were more and more differ- 
ent from those of its present condition. Now perhaps one of its 
component tendencies was relatively stronger, then another. But 
whatever these temporary slight fluctuations, it is certain that if 
we only go back far enough we must arrive at a stage where the 
tendencies were even more numerously and more intimately 
combined than now. But if one should believe that Arapaho art 
can be explained, for instance, by the coventionalized realism 
theory, the realism being original and the conventionalization 
subsequent, he holds the view that at some time past this Arapaho 
art consisted of pictorial representations. This view is logically 
possible, but in reality it is absurd. This art could not have had 
so ideally simple a development that we could still trace its 
original condition, if it were very old. But if it, therefore, were 
comparatively recent in origin, there must until a certain time 
have been no art among the Arapaho, while at that moment it 
sprang up full-blown, not as a crude undifferentiated thing, but a 
highly-specialized pictorial art. Such an event would be ex- 
tremely remarkable, not to say marvelous, and more in need of 
an explanation than the phenomenon it explained. By isolating 
any tendency that we find in any art, we are led to imagine a 
purely ideal condition which not only could not have been the 
original state of the art, but is probably even more different from 
its original state than from its present known state. 

In short, it is impossible to determine the origin of any art 
whose history we do not know. 

Let us briefly consider the field of mythology. There have 
been numerous explanations of myths and several theories of the 
origin of all mythology. The principal of these theories are the 

What may be called the physical or science theory accounts 

330 AMERICAN ANTHROPOLOGIST [n. s., 3, 1901 

for myths by making them the outcome of a desire to explain 
natural phenomena. The shapes or colors of animals, the 
motion of sun and moon, the existence of the stars, strange geo- 
logic formations, such phenomena are supposed to have stimu- 
lated the wonder of primitive man so much that he made myths 
to explain them. 

The personification theory supposes that deities and other 
mythic characters, together with their actions, — in a w^ord, myth- 
ology — are personifications of natural phenomena. Phoebus, 
Indra, Agni, are said to have originated in personifications of the 
sun, heaven, and fire. The solar myth theories, and others of an 
analogous kind, belong here. 

The animistic theory says that there was originally a belief in 
soul, out of which arose the various systems of spirits and deities. 
It believes that myths originated from a state of the human mind to 
which all objects seemed equally endowed with human personality. 

These three theories are at bottom the same. 

What has been called the allegorical or ethical theory supposes 
myths to be allegorical inventions with a moral import. Miracu- 
lous stories of gods, men, and animals are thought to have been 
composed in order to teach, by illustration, ethical precepts. This 
view is not so much in favor now as formerly. 

The historical theory makes myths the distortion of actual 
events. A powerful king of Crete gave rise to the mythic 
character of Zeus. 

The etymological theory calls mythology a disease of language. 
Misinterpreted metaphors or false etymologies gave rise to myths. 
To use a familiar example, Zeus is thought to have been originally 
called Kronion, with the meaning ** existing through all time." 
Later this epithet was misunderstood to mean son of Kronos, 
and thus gave rise to the conception of a god Kronos.* 

As explanations, all these theories are untrue. But the ten- 
dencies which they recognize exist. 

^ This does not necessarily exhaust the number of theories. 

kroeber] decora TIVE SYMBOLISM OF THE ARAPAHO 33 1 

There is undoubtedly a tendency to explain natural phenomena 
in myths. The Indians of British Columbia have this story : The 
bear and the chipmunk disputed whether there was to be dark- 
ness or light. The chipmunk triumphed, and for the first time it 
became light. The angry bear attacked the chipmunk and pur- 
sued it. The chipmunk escaped by tearing itself from under the 
claws of the bear. From this it is striped down its back. This 
little story, whatever its origin, clearly reflects the tendency to 
mythologize about such natural phenomena as day and night and 
the color-markings of animals. Hundreds of similar myths con- 
cerned with the spots on the moon, or the blackness of the crow, 
or a certain peculiar stone, or a similar fact, are known from all 
parts of the world. 

There is also a tendency to identify mythic personages with 
parts of nature ; Thor with thunder, for instance. And the ten- 
dency toward animism is so widespread and so deep-seated that it 
will be recognized without an example. 

It must also be admitted that there is something of an ethical 
tendency in mythologies. Among primitive races ceremonial and 
ritual partly take the place of our later morality. And very fre- 
quently myths deal with ceremonial. The American Indians, the 
Jews, the Australians, and the Greeks have such myths. 

The existence of a historical tendency in myths is demon- 
strated by the introduction of Attila into the Sigurd saga. 

The etymological tendency, finally, is revealed in the following 
extract from a Dakota myth * : An old couple have adopted a 
foundling. When he grows up he is so successful in killing 
buffalo that he makes his parents very rich in dried meat. 
"Then the old man said: * Old woman, I am glad we are well off. 
I will proclaim it abroad.' And so when the morning came he went 
up to the top of the house and sat, and said, * I, I have abundance 
laid up. The fat of the big guts {tashiyakd) I chew.* And 
they say that was the origin of the meadow-lark {tashiyakapopo) 

* Riggs, Z?df^<?/fl Grammar^ Texts ^ and Ethnography^ 1893. 

332 AMERICAN ANTHROPOLOGIST [n. s., 3, iqoi 

It has a yellow breast, and black in the middle, which is the 
yellow of the morning, and they say the black stripe is made 
by a smooth buffalo horn worn for a necklace.'* — From this point 
the myth deals with the adventures of the boy. 

It is thus clear that for every one of these theories there really 
exists a tendency in primitive man which influences his myths. 

This multiplicity of tendencies or causative forces necessarily 
refutes any explanation that uses and allows only one of them. 
Such have been all explanations of myths. Such they must be, 
for when more than one tendency or cause is admitted, we can have 
several tentative suggestions but no longer one positive explana- 
tion. The case is analogous to that in art, and does not require de- 
tailed restatement. It may be said, in short, that all explanations 
of myths consist of the ignoring of all the eternal and indestructible 
tendencies in man with the exception of one which is isolated and 
elevated as the sole cause of the myth. That such explanations, 
however clear and impressive they are, cannot be true, is obvious. 

Thus we come to the conclusion that all search for origins in 
anthropology can lead to nothing but false results. The tenden- 
cies of which we have spoken are at the root of all anthropological 
phenomena. Therefore it is these general tendencies more 
properly than the supposed causes of detached phenomena that 
should be the aim of investigation. 

These tendencies, being inherent in mind,* are everlasting. On 

' The tendencies spoken of throughout this essay must be understood to be the 
tendencies of social man. They are those tendencies which exist in individuals being 
parts of a culture, not in isolated individuals as such. There are psychological causes 
or mental conditions — generally considered physiological — which might also be called 
tendencies. Such are the tendency to fatigue, the tendency to form habits, the ten- 
dency toward imitation by suggestion, and others. These exist nearly identically in 
all men, whatever their degree of civilization ; they seem even to occur with little 
modification in animals. It is evident that these physiological tendencies are totally 
independent of cultures. Our knowledge concerning them is due to a psychological 
study of individual men. On the other hand those tendencies which alone are referred 
to above are determinable only from a historical study of social groups. The mani- 
festations of these tendencies are activities such as mythology, writing, ceremonials 
decorative art, castes, commerce, and language. 

kroeber] decora TIVE SYMBOLISM OF THE ARAPAHO 333 

the other hand they are constantly changing and developing, and 
varying in their differentiations and combinations. The phe- 
nomena of activity have changed as these tendencies and their 
relations to one another have become modified. Therefore the 
products of mind (the phenomena studied by anthropologists) 
are, like mind itself, beginningless (for us). They have no origin. 
All arts and all institutions are as old as man. Every word is as 
old as speech. The history of every myth is at least as long as 
the history of mankind. Of course no myth was ever alike from 
one generation to the next ; no decorative style has ever remained 
unaltered. But no myth, no artistic convention, nor any other 
thing human, ever sprang up from nothing. It always grew from 
something previous that was similar. These principles are obvi- 
ous, but they are ignored and implicitly denied in every search 
for an origin. , 

Every explanation of an origin in anthropology is based on 
three processes of thought which are unobjectionable logically 
but are contrary to evolutionary principles and the countless 
body of facts that support these principles. First is the assump- 
tion, implied in the word origin^ that before the beginning of the 
phenomenon explained, itself and its cause were absent ; second 
is the belief that a suddenly arising cause singly produced the 
phenomenon ; and the third is the idea that this cause as sud- 
denly and completely ceased as it had before sprung up, and that 
its product has remained, unaffected by other causes, unaltered 
but for wear and tear, to the present day. These three thought- 
processes are present in every explanation of the cause or origin 
of a human phenomenon, whether the explainer himself be con- 
scious or unconscious of them. Generally, indeed, the origin is 
not stated unhesitatingly and clearly enough for these three steps 
of thought to be visible in all their baldness. Often, perhaps, 
the investigator advancing a theory of origin would himself deny 
these processes to exist in his reasoning. Nevertheless, every de- 
termination of an origin, whether origin means the beginning of 

334 AMERICAN ANTHROPOLOGIST [n. s., 3, 1901 

a phenomenon or its cause, must imply the existence of, first, a 
previous different state, secondly, a change produced by an ex- 
ternal (non-inherent) cause, and, thirdly, the state that is being 

This three-step process of reasoning is not in itself wrong. 
When it is declared either that steam in a particular case was, 
or in general can be, produced from water by heat, this method of 
thought is employed. The early state is the water, the altering 
cause the heat, and the present state the steam. In all the phy- 
sical sciences thinking in this manner is not only permissible but 
necessary and is constantly done. It is when these thought-pro- 
cesses are used in anthropology* that their results become absurd. 
When we say that the origin of decoration is technique, or that the 
origin of marriage is promiscuity, or that the origin of the Polyne- 
sian Maui is personification of the sun, or that the origin of an al- 
phabet is pictorial art, or that the beginning or cause of anything 
in human culture is a certain other thing — we assert or imply a 
distinct and separate antecedent condition and an isolated, defin- 
itely limited efficient cause. That such a condition and such a 
cause really existed we have shown in the consideration of primi- 
tive art to be so highly improbable as to make the belief in their 
reality absurd ; and it must be obvious that in all other cases 
within the scope of anthropology the three suppositions made in 
every explanation of origin where direct historical knowledge is 
lacking, possess the same degree of improbability.* 

If, then, the specific causes or beginnings of specific phenomena 

* By the term anthropology there are meant here not those portions of the science 
which are clearly anatomical and physiological (i. e., resting upon mechanical science 
and included in it), but those domains generally covered by the titles ethnology, arche- 
ology, and history. 

^ If it is true that origins cannot be determined, the supposed origins of words, 
namely roots, must be imaginary. Whoever gives adherence to the currently ac- 
cepted theory that language began with roots, deliberately or unconsciously commits 
himself to these beliefs : That previous to the making of roots, language in the 
proper sense, as something articulate and definite, was wanting. That with the 
roots, language began to be, essentially as it is now. That after the formation of 
the roots no new ones ever arose, but language remained unchanged except for mod- 


kroeber] decora TIVE SYMBOLISM OF THE ARAPAHO 335 

are a delusion in anthropology and may not be sought, what can 
be the subject of investigation ? The tendencies that have been 
referred to so much ? Like words and styles and myths and ideas 
and industrial processes and institutions, all of which are their 
products, tendencies are both eternally living and everlastingly 
changing. They flow into one another ; they transform them- 
selves ; they are indistinguishably combined where they coexist. 
So, if our view is wide enough, we cannot properly determine and 
separate and name and classify tendencies. They really exist 
only in the whole unity of living activity as parts in the endless 
organism. This great unity is the true study for the student of 
man. In it, as parts of it, cultures and civilization-movements, 
tendencies and individual phenomena, are comprehensible. In it 
we know their interrelations. Only by understanding its totality 
can we really understand its smaller parts, those productions that 
have always a predecessor but never a beginning. 

The fundamental error of the common anthropological 
method of investigating origins is that it isolates phenomena and 
seeks isolated specific causes for them. In reality, ethnic phenom- 
ena do not exist separately : they have their being only in a cul- 
ture. Much less can the causative forces of the human mind, the 
activities or tendencies, be truly isolated. Every distinction of 
them is not only arbitrary but untrue. Both phenomena and 

ifications of its roots or their combinations into new words and inflectional forms. 
The improbability of such a process having ever taken place must be clear to any one 
who believes that never-dying, ever-changing, interrelated tendencies have unceas- 
ingly and unitedly been operative in man. The belief in roots as the sources of lan- 
guages is totally unevolutionary : it is contrary to the axiom that nothing living ever 
comes but from what is similar and that all change is gradual development and not a 
process of finished creation. The weakness of the theory of roots is most palpable in 
the absurdity of the various explanations that are frequently given of the origin of the 
primary roots. It is true that there is something that may be called roots. In every 
language there are groups of words similar in sound and related in meaning. The 
ideal, non-existent centers of these groups of words can well be named roots, and 
they must be recognized and used in philology. But roots that once existed as such, 
and gave rise to languages of words in which they can still be seen, — such there never 

33^ AMERICAN ANTHROPOLOGIST [n. s., 3, 1901 

causes can be properly apperceived only in the degree that we 
know their relations to the rest of the great unity that is called 
life. The more this is known and understood as a whole, the 
more do we comprehend its parts. This, the whole of life, is the 
only profitable subject of study for anthropology. 

Columbia University, 1 892-1 901. 


A.B., Columbia, 1896. 
A.M., Columbia, 1897. 


Animal Tales of the Eskimo. 
The Eskimo of Smith Sound. 
Symbolism of the Arapaho Indians. 
Tales of the Smith Sound Eskimo. 
Cheyenne Tales. 



tShc ftnlckerlMChcr recti* lUw Vor» 

• * S*- .'' _-" - - r-«i.